Younger mothers engage in higher rates of substance use and HIV risk behaviors, and their children are also more likely to engage in these behaviors and experience teenage pregnancy. Early unplanned pregnancies suggest that offspring of younger mothers are engaging in unprotected sex during adolescence, placing themselves at risk for STDs. No studies have compared maternal patterns of substance use before and during childrearing to examine the effect of maternal age on developmental patterns of substance use in isolation from other confounds associated with maternal age. The goal of this study is to examine trajectories of maternal substance use as a moderator of the effect of maternal age on child risky behavior and health. The proposed study will take advantage of 2 large existing community-based birth cohorts. In one, the mothers were 12-18 years old at the first prenatal assessment. In the other, the mothers were 18-42 years old. The aims of the original studies were to examine the long-term effects of prenatal substance use. Both cohorts include data on maternal substance use prior to pregnancy, during pregnancy, and 4 time points after pregnancy (when the children were 6, 10, 14 and 16 years old). The parent studies collected data from a wide variety of domains, including detailed demographic, psychological and behavioral profiles. The larger combined dataset will permit advanced and complex statistical models of substance use over time, and permit us to test several new questions about maternal age, substance use and child health stratified by race and gender. Our specific aims are to (1) identify maternal trajectories of substance use; (2) determine if maternal age at first birth predicts trajectory membership; (3) examine racial differences in substance use and reproductive risk; (4) investigate the effect of maternal substance use trajectories on reproductive risk in adolescent offspring of different races; and (5) test if maternal substance use moderates the effect of maternal age on reproductive risk in adolescent offspring of different races. An innovative feature of this application is that we are proposing and testing a new theoretical model of maternal age and maternal substance use on early reproductive risk in offspring. This model is based on vulnerability models and life course theory, and will help elucidate the mechanisms linking maternal age to risky behavior and reproductive health in the next generation. This model will provide empirical support for targeted intervention strategies with at-risk families. The proposal is both time- and cost-effective because we have access to 16 years of prospective, longitudinal data from two birth cohorts. To date, no studies have examined the role maternal age and long-term patterns of substance use in childbearing women to determine how maternal substance use may contribute to early child substance use and adolescent reproductive risk.